Synthesis and theoretical analysis of palladium complexes of polydimethylsiloxane functionalised pyridine and their catalytic activity in alcohol oxidations under low polar conditions

Article abstract of DOI:10.1016/j.poly.2010.09.004

Palladium(II) complexes of polydimethylsiloxane (PDMS)-functionalised pyridine (1), [Pd(OAc)₂(1)₂] (2) and [PdCl ₂(1)₂] (3), were synthesised and characterised. The cis/trans isomerism of models of complex 3 was investigated using DFT calculations. Models of complex 2 were also theoretically analysed, to identify any possible C-H?O interactions. Complex 2 was investigated as a catalyst for selective alcohol oxidations both under solventless conditions and in supercritical carbon dioxide (scCO₂). In scCO₂ the presence of 1 was found to stabilise the palladium complex, inhibiting its decomposition to Pd(0), a stabilisation not witnessed with non-PDMS functionalised pyridine analogues.

Full text of DOI:10.1016/j.poly.2010.09.004

Polyhedron 29 (2010) 3287–3293
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Synthesis and theoretical analysis of palladium complexes
of polydimethylsiloxane functionalised pyridine and their catalytic
activity in alcohol oxidations under low polar conditions
⇑
⇑
Matthew Herbert, Francisco Montilla , Agustín Galindo
Departamento de Química Inorgánica, Universidad de Sevilla, Aptdo 1203, 41071 Sevilla, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 28 May 2010
Accepted 5 September 2010
Available online 21 September 2010
Palladium(II) complexes of polydimethylsiloxane (PDMS)-functionalised pyridine (1), [Pd(OAc)2(1)2] (2)
and [PdCl2(1)2] (3), were synthesised and characterised. The cis/trans isomerism of models of complex
3 was investigated using DFT calculations. Models of complex 2 were also theoretically analysed, to iden-
tify any possible C–HÁ Á ÁO interactions. Complex 2 was investigated as a catalyst for selective alcohol oxi-
dations both under solventless conditions and in supercritical carbon dioxide (scCO2). In scCO2 the
presence of 1 was found to stabilise the palladium complex, inhibiting its decomposition to Pd(0), a sta-
bilisation not witnessed with non-PDMS functionalised pyridine analogues.
Keywords:
Palladium
Oxidation
Alcohol
Ó 2010 Elsevier Ltd. All rights reserved.
Homogeneous catalysis
Supercritical carbon dioxide
1. Introduction
carbosilane dendron groups [10d] producing soluble metal com-
plexes that were active catalysts in scCO2.
The use of solubilising polymer substituents in catalysis is a
strategy with numerous precedents [1]. However, the incorpora-
tion of light polysiloxanes, such as polydimethylsiloxanes (PDMS),
into catalyst structures to facilitate catalysis in non-polar media
was not reported until relatively recently [2]. PDMS is chemically
and thermally very stable, non-volatile and possesses high gas per-
meability. Besides these properties, these polymers display rela-
tively good solubility in nearly all organic solvents including
even very non-polar media such as scCO2 [3], which due to its
advantageous properties, is attracting increasing interest as a med-
ia for homogeneous catalysis [4]. However, scCO2 shows generally
poor solvent properties and is very non-polar [5], which often com-
plicates its use as a reaction solvent. The solubilities of reaction
components, in particular transition metal catalyst complexes,
are often too low to permit any kind of efficient reaction. In this re-
gard, the past few years have now seen a number of publications
on PDMS solubilisation in catalytic systems [6] including incorpo-
ration of Jacobsen [7] and Grubbs [8] catalysts in PDMS mem-
branes. The functionalisation of phosphine ligands with PDMS
has been used to solubilise a metal complex in scCO2, facilitating
homogeneous catalysis [9]. Similarly our own research group has
described ligand functionalisation with trimethylsilyl [10a–c] and
Additionally, some of our more recent work has begun to inves-
tigate the use of PDMS functionalised ligands in the solubilisation of
catalyst complexes in scCO2. This work has centred on our investi-
gations of the PDMS functionalised pyridine ligand 4-(polydimeth-
ylsiloxanyl-ethyl)pyridine (1) (Scheme 1) and we have prepared
complexes of rhenium [11a], copper [11b,c] and molybdenum
[11d] incorporating 1 which showed novel catalytic properties.
A wide range of catalytic organic transformations facilitated by
palladium compounds have been developed and extensively stud-
ied [12]. PDMS functionalised complexes with high solubilities in
very non-polar media might therefore have applications in a range
of catalytic processes and indeed, while this work was in progress,
Mizoroki-Heck reactions catalysed by soluble polysiloxane-sup-
ported palladium catalysts were described [13]. Consequently, the
preparation of palladium(II) complexes of ligand 1 is reported here,
including our observations and some results from catalytic studies
of alcohol oxidation in apolar media. Due to the capacity of scCO2 to
mix perfectly with gaseous reagents, such as dioxygen, there are
obvious additional advantages to its use in this reaction [14].
2. Results and discussion
2.1. Synthesis of PDMS functionalised palladium(II) complexes
⇑
Palladium(II) acetate [15] possesses good solubility in the
majority of organic solvents, including halocarbons and toluene.
Corresponding authors. Fax: +34 954 557153.
E-mail address: galindo@us.es (A. Galindo).
0277-5387/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2010.09.004

3288
M. Herbert et al. / Polyhedron 29 (2010) 3287–3293
2.2. Theoretical analyses of complexes 2, 3, [PdCl2(py)2] and
[PdCl2(4VP)2] (4VP = 4-vinylpyridine)
Si
C2H4
N
O
n
In a previous work, trans/cis isomerism in related [PdCl2(P)2]
complexes, with P representing a phosphine species with a carbos-
ilane dendron function, was analysed [10d]. A similar theoretical
study has now been carried out for the complexes discussed here.
Scheme 2 shows the trans-configuration for complexes 2 and 3.
Experimentally this is the most stable configuration that
[PdCl2(N)2] compounds can adopt (N representing a monodentate
N-donor ligand). In some cases, the cis isomer is isolable, but will
convert to the thermodynamically favourable trans isomer when
heated. All cis-dichloropalladium complexes with pyridine-type
coligands to have been X-ray characterised have a bidentate ligand
(usually bipyridine or related) which forces the geometry.1 To con-
firm that the trans isomer is the most stable for complexes 2 and 3,
a theoretical DFT type analysis was conducted for several model
compounds representing complexes 2 and 3 (Scheme 3). The mod-
el compounds possessed two p-ethyl-PDMS (b isomer2) functional-
ised pyridines, the shortened (n = 2 or 4) PDMS functions
terminating with Si(CH3)3, (NC5H4-CH2CH2-(SiMe2O)n-Si(CH3)3).
The [PdCl2(py)2] and [PdCl2(4VP)2] complexes were also computed
for comparative purposes.
The DFT study began by investigating the chloride complexes
with shorter PDMS chains (n = 2), model 32. The optimised struc-
tures calculated for trans-32 and cis-32 are shown in Fig. S3 (Sup-
plementary data). Energetically, the trans isomer was calculated
to be more stable than the cis by 10.0 kcal molÀ1 (electronic en-
ergy). In order to evaluate the importance of steric factors in this
difference, calculations were carried out for the analogous cis-
and trans-isomers of [PdCl2(py)2] and [PdCl2(4VP)2] using the
same computational level. Fig. S4 shows the optimised struc-
tures. Roughly the same energy difference was computed be-
tween the two isomers (ca. 12 kcal molÀ1). This indicates that
steric influences from the PDMS chains have no significant role
in stabilising the trans isomer with respect the cis. Calculations
performed for the complex with longer PDMS chains (n = 4),
model 34, again found the trans isomer to be more stable than
the cis by 10.9 kcal molÀ1. The optimised structures calculated
for trans-34 and cis-34 are shown in Fig. 1. The PDMS substitu-
ents, located relatively far away from the metal centre, have a
high degree of conformational freedom and thus can easily adopt
conformations which minimise steric repulsions, even in the cis
isomer.
Scheme 1. Polydimethylsiloxane functionalised pyridine 1.
However, its solubility in highly apolar media, such as hexane is
demonstrably poor. Preparation of the complex of ligand 1 was
therefore interesting, since the PDMS functionalisation would be
anticipated to improve the solubility in this type of media. 4-(Poly-
dimethylsiloxanyl-ethyl)pyridine (1) was prepared by the
hydrosilylation of 4-vinylpyridine with hydride terminated poly-
dimethylsiloxane catalysed by Karstedt’s catalyst, following the
procedure previously described by us [11]. Preliminary spectro-
scopic data and a more detailed characterisation of 1 have already
appeared and therefore this will not be discussed again here [11].
Ligand 1 consists of a 2:1 mixture of the b and
a addition products
of the hydrosilylation reaction. Reaction of palladium(II) acetate
with 1 yields the corresponding coordination complex of diacetat-
opalladium, 2. This compound was characterised as [Pd(OAc)2(1)2]
(Scheme 2), taking the form of a yellow oil. The product workup re-
quires extraction with hexane, confirming solubility of the product
in such low polar solvents. Analysis of the 1H NMR and 13C{1H}
NMR spectra of 2 demonstrated that complexes with a low n value
(n = the number of PDMS monomer units {Si(CH3)2-O}n per termi-
nal pyridine) and those of the
a-addition isomer of 1 had been
eliminated during the product workup. In the 1H NMR spectrum
of 2 shown in Fig. S1 (Supplementary data), only signals corre-
sponding to the b-addition isomer of 1 are present, the CHCH3
peaks observable at 1.29 and 1.85 ppm in the spectrum of 1 now
absent, with only one peak each for the protons of the pyridine
ring, indicating the presence of only one isomer. Similarly in the
13C{1H} NMR spectrum (see Fig. S2), only signals corresponding
to the b-addition isomer are visible. From the peak integrals of
the 1H NMR spectrum the value of n was calculated to be 13, whilst
n ꢀ 5 had been calculated for the sample of 1 from which 2 was
prepared, indicating that the part of 2 formed from fractions of 1
with lower n were eliminated during the workup, presumably
due to poorer solubility in the very low polar hexane used to ex-
tract the product. Evidence supporting this was obtained in the
synthesis of the dichloro analogue, 3, discussed below, and similar
b isomeric purification and n increase was observed in the synthe-
sis of methyltrioxorhenium analogues [11a].
Pure dichloropalladium adopts a polymeric structure with l2
-
The complex subsequently studied in catalysis (see below) was
the acetate derivative 2, and the theoretical study was thus ex-
tended to the models trans-22 and trans-24. Fig. 2 shows the corre-
sponding optimised structures. Previously, theoretical studies
carried out for model methyltrioxorhenium (MTO) complexes of
the PDMS functionalised pyridine, showed C–HÁ Á ÁO interactions
between the PDMS chains and the oxo groups of the rhenium
[11a]. These interactions likely play an important role in stabilising
the catalyst complex during epoxidative catalysis, leading to signif-
icantly enhanced TONs. In the palladium acetate models 22 and 24,
however, no such interaction involving the PDMS chains was ob-
servable in proximity to the metal centre. Consequently, the higher
stability of 2 under the catalytic conditions, compared with its
unsubstituted analogue (see discussions below), is unlikely to be
attributable to any C–HÁ Á ÁO type interaction involving the PDMS
chain.
chloride bridges and displays very poor solubility in most media;
typically it is dissolved by refluxing in a coordinating solvent such
as acetonitrile or benzonitrile forming the corresponding monome-
tallic coordination complex which can then undergo substitution
reactions. The synthesis of the dichloropalladium complex of 1,
namely 3, was achieved by heating the two components at 80 °C
in ethanol overnight, followed by hexane extractions of the
product. Complex 3 was obtained as a yellow oil and identified
as [PdCl2(1)2] (Scheme 2). Similar to the diacetato counterpart 2,
only complexes formed from the b-addition complex of 1 were
present in the final isolated sample of 3, there was no observable
trace of the a-addition product of 1. Also in common with 2, 3 pos-
sessed a far higher value of n, approximately 15, than the sample of
1 from which it was prepared (again n ꢀ 5). It was assumed that 3
complexes formed from the lower n fractions of 1 were being elim-
inated during the workup, due to their insolubility in hexane. Evi-
dence that this was the case was obtained from 1H NMR analysis of
the last solid fraction from which 3 was separated. Examination of
the integrals showed an n value of 4, indicating that the lower
polymer lengths in this sample were insufficient for solubilisation
in hexane.
1
CSD Search. Cambridge Structural Database System, Cambridge Crystallographic
data Centre, 12 Union Road, Cambridge, CB2 1EZ, UK.
2
Calculations of the
a and b isomers of ligand 1 appeared in Ref. [11a].

M. Herbert et al. / Polyhedron 29 (2010) 3287–3293
3289
Si
C2H4
N
O
X
n
1
Si
N
Pd
X
N
Si
C2H4
C2H4
PdX2
O
O
n
n
X
= AcO, 2
= Cl, 3
Scheme 2. Synthesis of palladium complexes with PDMS functionalised pyridine ligands.
X
CH2CH2
n
N
Pd
X
N
CH2CH2 Si
Si
n
Si
Si
O
O
X
CH2CH2
n
N
Pd
N
X
Si
Si
O
X
CH2
CH2
OAc Cl
n
22
24
32
34
2
Si
4
O
n
Si
Scheme 3. Models used in the DFT study.
2.3. Palladium catalysed aerobic alcohol oxidation in apolar media
however, are quite capable of facilitating the oxidation mechanism
and thus the diacetatopalladium complex 2 has potential as a cat-
alyst for this process. The principle objective of the ensuing cata-
lytic study was to demonstrate advantages stemming from the
solubility of the catalyst in very non-polar media, which might per-
mit homogeneous catalysis under conditions where it would be
impossible to use a non-PDMS functionalised analogue to obtain
decent product yields. The activity of 2 in the catalytic oxidation
of alcohols to their corresponding carbonyl products was thus
investigated both in solventless conditions and in scCO2.
The majority of palladium(II) catalysed homogeneous alcohol
oxidation systems to have been reported, including those of Peter-
son and Larcock [16], Uemura et al. [17a,b], and Sigman et al. [18],
typically employed organic solvents (e.g. DMSO, toluene and THF)
as the reaction medium. Sheldon et al. introduced an aqueous sys-
tem, employing a palladium catalyst with ligands possessing ionic
functional groups, that was soluble in the highly polar reaction
media [19]. This system is highly effective in the conversion of
smaller more polar alcohols with sufficient solubilities in water,
which form a biphasic reaction system with the ‘‘green” aqueous
reaction solvent. This system is a good example of a catalyst being
designed to function well in a specific reaction environment. In a
similar vein the work described here aimed to facilitate active
catalysis in very apolar media, such as ‘‘green” scCO2 and solvent-
less conditions. Having developed palladium complexes with good
solubilities in low polar media, their catalytic efficacy in the selec-
tive oxidation of alcohol substrates in such reaction conditions was
thus examined.
2.4. Oxidation of 2-octanol under solventless conditions
In a previous publication we presented a highly efficient sol-
ventless epoxidation system employing ligand 1 in conjunction
with methyltrioxorhenium as catalyst and we hoped to observe a
similar enhancement using palladium in alcohol oxidations [11a].
Since the objective of this study was to demonstrate whether 2
would display enhanced activity compared to non-PDMS function-
alised analogues in very low polar conditions, a low polar alcohol
substrate, 2-octanol, was chosen for the study of activity in a sol-
ventless system. A more polar substrate, e.g. benzyl alcohol, would
likely dissolve unfunctionalised analogues of 2 very efficiently,
with there thus being no necessity to enhance the solubility of
the catalyst complex. Under some standard conditions [17a,b] a
study of the solventless oxidation of this substrate catalysed by
The oxidation mechanism requires deprotonation of the alcohol
substrate, a step which cannot be catalysed by non-specialised di-
chloro complexes, rendering them inactive as catalysts in this par-
ticular transformation [17c]. The dichlorocomplex
3 would
therefore not be expected to be catalytically active in this transfor-
mation, and thus it was not investigated. Carboxylate complexes,

3290
M. Herbert et al. / Polyhedron 29 (2010) 3287–3293
Fig. 1. Optimised structures of model compounds trans-34 and cis-34.
Fig. 2. Optimised structures of model compounds trans-22 and trans-24.
[Pd(OAc)2] was conducted, examining the influence of both
unfunctionalised pyridine and 1 on the reaction. Dioxygen was
used as oxidant. The results are shown in Table 1.
Both base species were found to inhibit the rate of the oxidation
reaction (compare entries 2 and 3 with entry 1, Table 1). When no
base species was employed a markedly higher yield was obtained
whilst pyridine and 1 both reduced the level of conversion to
around a third. This is in common with observations made even
in early examples of Pd(II) catalysed alcohol oxidations where
coordinating bases were observed to prevent catalyst decomposi-

M. Herbert et al. / Polyhedron 29 (2010) 3287–3293
3291
Table 1
tem. A brief study investigating the catalytic oxidation of benzyl
alcohol and 2-octanol by the PDMS functionalised palladium(II)
acetate catalyst and some unfunctionalised analogue complexes
in scCO2 was thus conducted. The results are presented in Table 2.
In the oxidation of benzyl alcohol (entries 1–4, Table 2) only
negligible differences in the catalytic activity of the PDMS func-
tionalised catalyst (entries 2 and 4, Table 2) and the unfunctiona-
lised analogue complex of 4-vinylpyridine (entries 1 and 3, Table
2) were observed, regardless of whether the catalyst complex
was previously synthesised or allowed to form in situ. This latter
observation would seem to indicate that the coordination complex
of palladium acetate and the ligand species formed fairly quickly in
the reactor. In the oxidation of 2-octanol (entries 5 and 6, Table 2),
when unfunctionalised pyridine was used a limited yield was
recovered (entry 5, Table 2). However, when 1 was used only trace
conversion to the product was witnessed (entry 6, Table 2).
From the results it can quickly be concluded that in no case did
1 confer any advantage to the catalytic system through dissolution
of the palladium catalyst, as had been hoped. The study with re-
spect to catalytic applications was therefore concluded. It is inter-
esting, however, to make note of the state of the palladium catalyst
following the oxidation in each case. When unsubstituted pyri-
dines were employed the catalyst was completely reduced to pal-
ladium black, deposited onto all of the internal surfaces of the
reactor (see vial (a) in Fig. S6, Supplementary data). In contrast,
when 1 was employed the palladium apparently largely survived
as a Pd(II) complex, coating the sides of the reactor as the oily or-
ange coordination compound (see vial (b) in Fig. S6). This indicated
that under the reaction conditions 2 possessed greater stability
than its unsubstituted analogue. In analogous fashion, in diacetat-
opalladium catalysed alcohol oxidations in standard organic med-
ia, coordinating bases such as DMSO [22] and pyridine [20] are
observed to prevent catalyst decomposition by stabilising interme-
diate Pd(0) complexes in solution, preventing coagulation as palla-
dium black nanoparticles.
Solventless oxidation of 2-octanol.a
Entry
Ligand
Alcohol substrate
Yield (%)
1
2
3
–
2-Octanol
2-Octanol
2-Octanol
69
28
23
Pyridine
1
a
treaction = 18 h, T = 80 °C, alcohol substrate (5.0 mmol), [Pd(OAc)2] (0.05 mmol,
1%), ligand (0.25 mmol, 5%), and O2 (1 bar).
tion but also reduce the reaction rate. There was little difference in
the yields obtained using 1 (entry 3, Table 1) or unfunctionalised
pyridine (entry 2, Table 1), seemingly indicating that using the for-
mer did not result in any enhancement of the catalytic system.
However, in the reactions which employed base species, the cata-
lyst was observed to retain its yellow orange colour at the end of
the reaction time, whilst in the reaction run in the absence of base
the palladium had apparently converted completely to palladium
black. These observations are in common with those regularly ob-
served in Pd(II) catalysed alcohol oxidations [17a,b,20], where
coordinating bases inhibit the mechanism of the oxidation by com-
peting with the substrate for coordination sites, but also inhibit
decomposition of the catalyst by preventing agglomeration of the
Pd(0) complex which forms during the catalytic cycle. There are
two possible reasons for the higher yield observed here when no
base species was employed. It may be that the oxidation by the
[Pd(OAc)2] proceeded very quickly for a short period with a rela-
tively high yield produced before the catalyst was deactivated by
its reduction to palladium black. In this case the less active catalyst
species were not able to reach a higher level of conversion during
the reaction time despite the greater stability and thus much high-
er catalyst lifetime in these systems. The other possibility is that
the reaction was actually efficiently catalysed by the palladium
black formed from the catalyst decomposition when no base was
employed and that this catalytic system was actually more active
than the [Pd(OAc)2]-base systems [21].
3. Experimental
2.5. Oxidation of 2-octanol in scCO2
3.1. General
Due to the desirable sustainability credentials of the solvent,
alcohol oxidations which utilised scCO2 as the reaction medium
would present advantages from a green perspective. Additionally,
with gaseous O2 as the oxidant, improved reaction rates and selec-
tivities might be realised due to the perfect miscibility of the oxi-
dant with scCO2. Our previous studies with copper complexes
demonstrated that PDMS substituents can confer solubility in
scCO2 to a catalyst precursor [11b,c] and a similar solubilisation
was considered possible for the palladium complexes. Diacetat-
opalladium complexes of simple bases, i.e. pyridine, should possess
negligible solubilities in scCO2. However, the PDMS functionalisa-
tion of catalyst 2 apparently facilitates its dissolution to some ex-
tent (see Fig. S5 in Supplementary data), which is potentially
crucial in activating the catalyst in a homogeneous catalytic sys-
All preparations and other operations were carried out under
dry anaerobic conditions. Solvents were dried, using standard pro-
cedures. Palladium compounds and other chemicals were pur-
chased from Aldrich and were used as supplied. Infrared spectra
were recorded on Perkin–Elmer Model 883 spectrophotometer.
NMR spectra were obtained using a Bruker AMX-300 spectrometer.
13C{1H} and 1H shifts were referenced to the residual signals of
deuterated solvents. All data are reported in ppm downfield from
Me4Si. Elemental analysis (C, H, N) were performed on an Elemen-
tal LECO CHNS 93 analyser by the Microanalytical Service of the
Universidad de Sevilla (CITIUS).
3.2. Synthesis of trans-[Pd(OAc)2(4VP)2]
Working always under N2, two equivalents of 4-vinylpyridine
(4VP) (113.5 lL, 1.0 mmol) were added to a solution of palla-
Table 2
Oxidation of alcohols in scCO2.a
dium(II) diacetate (112.2 mg, 0.5 mmol) in Cl2CH2 (10 mL). Imme-
diate precipitation of the product as a yellow solid was observed
and the mixture was left to stir for 6 h. Subsequently the product
was separated by decantation and the solvent removed under vac-
uum yielding the crude product as a yellow powder. Recrystallisa-
tion from Cl2CH2 yielded [Pd(OAc)2(4VP)2] as a microcrystalline
yellow solid (145.7 mg, 0.335 mmol, 67% yield). IR (KBr, cmÀ1):
695, 853, 874, 925, 956, 995, 1012, 1033, 1076, 1306, 1377,
1454, 1598, 2853, 2923. 1H NMR (CDCl3): d 1.86 (s, 3H, CH3CO2),
Entry
Catalyst/ligand
Alcohol substrate
Yield (%)
1
2
3
4
5
6
[Pd(OAc)2]/4-vinylpyridine
[Pd(OAc)2]/1
[Pd(OAc)2(4VP)2]
2
[Pd(OAc)2]/pyridine
[Pd(OAc)2]/1
Benzyl alcohol
Benzyl alcohol
Benzyl alcohol
Benzyl alcohol
2-Octanol
30
40
30
35
26
2
2-Octanol
a
treaction = 18 h, T = 80 °C, P(CO2) = 150 bar, P(O2) = 1 bar, alcohol substrate
(5.0 mmol), [Pd] (0.05 mmol, 1%), and ligand (0.25 mmol, 5%).

3292
M. Herbert et al. / Polyhedron 29 (2010) 3287–3293
5.62 (m, 1H, Py-CHCHcisHtrans, cis/trans with respect to 4-pyridyl
group), 6.03 (m, 1H, Py-CHCHcisHtrans), 6.65 (m, 1H, Py-CHCH2),
7.29 (m, 2H, m-NC5H4), 8.61 (m, 2H, o-NC5H4). 13C{1H} NMR
(CDCl3): d 23.3 (CH3CO2), 121.7 (Py-CHCH2), 121.7 (m-NC5H4),
133.3 (Py-CHCH2), 147.2 (p-NC5H4), 151.5 (o-NC5H4), 178.1
(CH3CO2). Anal. Calc. for PdC18H20N2O4 (434.78): C, 19.72; H,
4.64; N, 6.44. Found: C, 19.29; H, 4.76; N, 6.44%.
03 suite of programs [27]. Coordinates of all optimised compounds
are collected in Supplementary data.
3.6. General experimental procedure for the 2-octanol oxidation
reactions
3.6.1. Solventless
Reactions were conducted in
a
glass Lab CrestÒ reactor
equipped with a high pressure valve. The specified quantities
(see footnote to Table 1) of catalyst and alcohol substrate were
charged and the reactor sealed. O2 (1 bar) was then applied and
the reactor heated at 80 °C with stirring for the duration of the
reaction. When the reaction had finished the reactor was cooled
in an ice bath and de-pressurised. The products were then dis-
solved in an appropriate solvent and analysed by GC.
3.3. Synthesis of trans-[Pd(OAc)2(1)2], (2)
A mixture of palladium(II) diacetate (50.0 mg, 0.22 mmol) and 1
(225 mg, 0.45 mmol) in dry acetone (10 mL) was stirred at room
temperature overnight. The reaction mixture was then evaporated
to dryness by application of vacuum. The resulting brown oil was
extracted with hexane (15 mL) affording a brownish solution,
which was left for a week at room temperature. A beige precipitate
was observed to precipitate, leaving a yellow solution which was
separated by decantation followed by centrifugation and evapo-
rated under vacuum to leave compound 2 as a yellow oil (40 mg,
8% yield in [Pd]). IR (NaCl, cmÀ1): 661, 696, 801, 863, 916, 1026,
1092, 1261, 1310, 1366, 1412, 1434, 1502, 1576, 1598, 1619,
2905, 2962, 3351. 1H NMR (CDCl3): d 0.06 (s, ꢁ156H, Si(CH3)2),
0.84 (t, JHH = 5.4 Hz, 4H, Py-CH2CH2-Si), 1.83 (s, 6H, CH3CO2), 2.65
(t, JHH = 5.3 Hz, 4H, Py-CH2CH2-Si), 7.13 (m, 4H, m-NC5H4), 8.51
(m, 4H, o-NC5H4). 13C{1H} NMR (CDCl3): d 0.00 (Si(CH3)2), 18.60
(Py-CH2CH2-Si), 23.18 (CH3CO2), 28.81 (Py-CH2CH2-Si), 124.30
(m-NC5H4), 150.99 (o-NC5H4), 157.41 (p-NC5H4), 178.14 (CH3CO2).
29Si{1H} NMR (CDCl3): d À21.97 (s, O-Si(CH3)2-O), À20.80 (s, CH2-
Si(CH3)2-O). Anal. Calc. PdC70H176N2O30Si26 (2362.78, formula for
n = 13): Pd, 4.50; C, 35.58; H, 7.51; N, 1.19. Found: Pd, 4.47; C,
35.13; H, 7.59; N, 1.35%.
3.6.2. In scCO2
Catalytic reactions in scCO2 were carried out in a in a 27 mL
stainless steel high-pressure cell. The 2-octanol substrate was
charged to the reactor along with a stirrer flea and the specified
quantity (see footnote to Table 2) of catalyst was then introduced
in a separate vial also containing a small stirrer flea, in this manner
preventing contact between the substrate and co-catalysts during
the reaction without dissolution occurring. The reactor was sealed
and air removed by application of vacuum. O2 (1 bar) was then
introduced, the reactor heated to 80 °C in a thermostatted oil bath
and CO2 was pumped until 150 bar was achieved. Stirring was then
initiated. After reaction the reactor was cooled in an ice bath and
de-pressurised through a solvent trap. The products were extracted
from the reactor with an appropriate solvent and the extract com-
bined with the contents of the solvent trap. The solution was then
analysed by GC.
3.4. Synthesis of trans-[PdCl2(1)2], (3)
4. Conclusions
A mixture of palladium(II) dichloride (52.0 mg, 0.293 mmol)
and 1 (300 mg, 0.59 mmol, 2 equiv.) in ethanol (15 mL) was stirred
at 80 °C overnight. The resulting mixture was then filtered to re-
move palladium black which had formed affording a yellow solu-
tion. The solvent was then evaporated by applying vacuum
leaving a turbid yellow oil. This was extracted with hexane with
the resulting yellow solution separated from the solid residues
by centrifugation and left for a week at room temperature. A brown
solid precipitated which was separated by centrifugation, the solu-
tion affords compound 2 in the form of a yellow gum (52.4 mg,
3.3% yield in [Pd]). IR (NaCl, cmÀ1): 659, 689, 703, 798, 863, 954,
1097, 1261, 1413, 1430, 1617, 2905, 2963. 1H NMR (CDCl3): d
0.09 (s, ꢁ180H, Si(CH3)2), 2.68 (t, JHH = 8.7 Hz, 4H, Py-CH2CH2-Si),
7.17 (m, 4H, m-NC5H4), 8.67 (m, 4H, o-NC5H4). Anal. Calc.
PdC74H194N2O30Cl2Si30 (2612.22): Pd, 4.07; C, 34.02; H, 7.49; N,
1.07. Found: Pd, 1.97; C, 36.01; H, 7.95; N, 0.99%.
In summary, the [Pd(OAc)2(1)2] (2) and [PdCl2(1)2] (3) com-
pounds have been synthesised via the reactions of the correspond-
ing palladium(II) precursor with polydimethylsiloxane (PDMS)-
functionalised pyridine (1), and adequately characterised. DFT
calculations have been performed with models of 2 and 3 in order
to investigate the cis/trans isomerism in such a complexes and
to identify any possible C–HÁ Á ÁO interactions in 2. Complex
[Pd(OAc)2(1)2] (2) has been investigated as a catalyst for selective
alcohol oxidations both under solventless conditions and in super-
critical carbon dioxide (scCO2). The PDMS functionalised pyridine
ligand 1 was thus shown not to confer significant advantages in
alcohol oxidations in the apolar environments of either solventless
or scCO2 reactions. However, the apparent stability of 2 compared
to unfunctionalised analogues under the reaction conditions may
indicate that in other palladium catalysed systems where 2 was
more active there could be some advantage to its use. With palla-
dium(II) catalysts being useful in a wide range of transformations,
not limited to oxidation reactions, there may well exist systems in
which the functionalised complexes 2 and 3 would represent ideal
specialised catalysts. Further studies in this direction are in
progress.
3.5. Computational details
The electronic structure and geometries of the model com-
pounds were computed by density functional theory at the
B3LYP level [23,24]. The Pd atom was described using the LANL2DZ
*
basis set [25,26], while the 6-31G basis set was used for the Cl, Si,
C, O, N and H atoms. The optimised geometries of the compounds
trans- and cis-[PdCl2(py)2] and trans- and cis-[PdCl2(4VP)2]
(4VP = 4-vinylpyridine) were characterised as energy minima by
the nonexistence of imaginary frequencies (NImag = 0) in the diag-
onalisation of the analytically computed Hessian (vibrational fre-
quencies calculations). For the larger systems, 22, 24, 32 and 34,
frequency calculations were considered too expensive and were
not carried out. DFT calculations were performed using the GAUSSIAN
Acknowledgements
This research was supported by the Spanish Ministerio de Cien-
cia e Innovación (CTQ2007-61037) and Junta de Andalucía (Proyec-
to de Excelencia, FQM-02474). We are grateful to Mr. Carlos
Carrasco who assisted in the catalytic studies and also to CICA
and UGRGrid (University of Granada) for permitting the use of
their computational resources.

M. Herbert et al. / Polyhedron 29 (2010) 3287–3293
3293
(d) F. Montilla, A. Galindo, R. Andrés, M. Córdoba, E. de Jesús, C. Bo,
Organometallics 25 (2006) 4138.
Appendix A. Supplementary data
[11] (a) M. Herbert, A. Galindo, F. Montilla, Organometallics 28 (2009) 2855;
(b) M. Herbert, F. Montilla, A. Galindo, Dalton Trans. 39 (2010) 900;
(c) M. Herbert, F. Montilla, A. Galindo, Inorg. Chem. Commun. 10 (2007) 735;
(d) M. Herbert, Ph.D. Thesis, Universidad de Sevilla, 2010.
[12] See for example: T. Nishimura, S. Uemura, Synlett 2 (2004) 201.
[13] M. Cypryk, P. Pospiech, K. Strzelec, K. Wasikowska, J.W. Sobczak, J. Mol. Catal. A
319 (2010) 30.
[14] See for example: T. Seki, A. Baiker, Chem. Rev. 109 (2009) 2409.
[15] A.C. Skapski, M.L. Smart, Chem. Commun. (1970) 658.
[16] K.P. Peterson, R.C. Larcock, J. Org. Chem. 63 (1998) 3185.
[17] (a) T. Nishimura, T. Onoue, K. Ohe, S. Uemura, Tetrahedron Lett. 39 (1998)
6011;
Spectra of 2, optimised structures of 32, [PdCl2(py)2] and
[PdCl2(4VP)2], solubilisation of 2 in scCO2, visual appearance of cat-
alysts after the oxidation reactions in scCO2 and coordinates of
optimised compounds.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.poly.2010.09.004.
References
(b) T. Nishimura, T Onoue, K. Ohe, S. Uemura, J. Org. Chem. 64 (1999) 6750;
(c) D.R. Jensen, M.J. Schultz, J.A. Mueller, M.S. Sigman, Angew. Chem., Int. Ed.
42 (2003) 3810.
[1] See for example: D.E. Bergbreiter, J. Tian, C. Hongfa, Chem. Rev. 109 (2009) 530.
[2] M.A. Grunlan, K.R. Regan, D.E. Bergbreiter, Chem. Commun. (2006) 1715.
[3] (a) M.A. McHugh, V.J. Krukonis, Supercritical Fluid Extraction: Principles and
Practice, Butterworth-Heinemann Ltd., Boston, 1994;
(b) V. Chandrasekhar, Inorganic and Organometallic Polymers, Springer, 2005
(Chapter 6).
[4] (a) See for example: W. Leitner, Acc. Chem. Res. 35 (2002) 746;
(b) J.L. Kendall, D.A. Canelas, J.L. Young, J.M. DeSimone, Chem. Rev. 99 (1999)
543;
[18] M.J. Schultz, C.C. Park, M.S. Sigman, Chem. Commun. (2002) 3034.
[19] G.-J.T. Brink, I.W.C.E. Arends, R.A. Sheldon, Science 287 (2000) 1636.
[20] B.A. Steinhoff, S.S. Stahl, Org. Lett. 4 (2002) 4179.
[21] (a) For examples of aerobic alcohol oxidation by Pd(0) nanoparticle catalysts
see the following: H. Wu, Q. Zhang, Y. Wang, Adv. Synth. Catal. 347 (2005)
1356;
(b) J. Chen, Q. Zhang, Y. Wang, H. Wan, Adv. Synth. Catal. 350 (2008) 453;
(c) Z. Hou, N. Theyssen, W. Leitner, Green Chem. 9 (2007) 127;
(d) N. Jamwal, M. Gupta, S. Paul, Green Chem. 10 (2008) 999;
(e) F. Li, Q. Zhang, Y. Wang, Appl. Catal. A: Gen. 334 (2008) 217. and relevant
references therein.
(c) P.J. Jessop, T. Ikariya, R. Noyori, Chem. Rev. 99 (1999) 475;
(d) P. Licence, J. Ke, M. Sokolova, S.K. Ross, M. Poliakoff, Green Chem. 5 (2003)
99;
(e) M.F. Kemmere, T. Meyer (Eds.), Supercritical Carbon Dioxide in Polymer
Reaction Engineering, Wiley-VCH, 2005.
[22] (a) S.S. Stahl, J.L. Thorman, R.C. Nelson, M.J. Kozee, Am. Chem. Soc. 123 (2001)
7188;
(b) B.A. Steinhoff, S.R. Fix, S.S. Stahl, J. Am. Chem. Soc. 124 (2002) 766.
[23] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[24] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[25] T.H. Dunning Jr., P.J. Hay, Modern Theoretical Chemistry, Plenum, New York,
976, p. 1.
[5] P. Raveendran, Y. Ikushima, S.L. Wallen, Acc. Chem. Res. 38 (2005) 478.
[6] (a) A.L. Miller II, N.B. Bowden, Chem. Commun. (2007) 2051;
(b) M.T. Mwangi, M.B. Runge, K.M. Hoak, M.D. Schulz, N.B. Bowden, Chemistry
14 (2008) 6780.
[7] (a) R.F. Parton, I.F.J. Vankelecom, D. Tas, K.B.M. Janssen, P.-P. Knops-Gerrits,
P.A. Jacobs, J. Mol. Catal. A 113 (1996) 283;
(b) K.B.M. Janssen, I. Laquiere, W. Dehaen, R.F. Parton, I.F.J. Vankelecom, P.A.
Jacobs, Tetrahedron: Asymmetr. 8 (1997) 3481;
[26] P.J. Hay, W.R. Wadt, J. Chem. Phys. 82 (1985) 299.
[27] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A. Montgomery, Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox,
H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E.
Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y.
Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S.
Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J.
Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L.
Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M.
Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A.
Pople, Gaussian 03, Revision C.02, Gaussian, Inc., Wallingford, CT, 2004.
(c) D.F.C. Guedes, T.C.O. MacLeod, M.C.A.F. Gotardo, M.A. Schiavon, I.V.P.
Yoshida, K.J. Ciuffi, M.D. Assis, Appl. Catal. A: Gen. 296 (2005) 120;
(d) J.R. Pliego Jr., M.A. Schiavon, J. Phys. Chem. C 112 (2008) 14830.
[8] (a) M.T. Mwangi, M.B. Runge, N.B. Bowden, J. Am. Chem. Soc. 128 (2006)
14434;
(b) M.T. Mwangi, M.B. Runge, N.B. Bowden, J. Organomet. Chem. 691 (2006)
5278.
[9] S. Saffarzadeh-Matin, C.J. Chuck, F.M. Kerton, C.M. Rayner, Organometallics 23
(2004) 5176.
[10] (a) M.F. Sellin, I. Bach, J.M. Webster, F. Montilla, V. Rosa, T. Avilés, M. Poliakoff,
D.J. Cole-Hamilton, J. Chem. Soc., Dalton Trans. (2002) 4569;
(b) F. Montilla, V. Rosa, C. Prevett, T. Avilés, M. Nunes Da Ponte, D. Masi, C.
Mealli, Dalton Trans. (2003) 2170;
(c) F. Montilla, A. Galindo, V. Rosa, T. Avilés, Dalton Trans. (2004) 2588;